An invention for identifying a spatial location of an event within video image data is provided. Disclosed are embodiments for detecting an object and obtaining trajectory data of a trajectory of the object within the video image data from a sensor device; converting the trajectory data into a contour-coded compressed image; generating, based on the trajectory data, a searchable code that contains a set of locations traversed by the trajectory of the object within the video image; associating the searchable code with the contour-coded compressed image in a database; and returning, in response to a query having a selected location that corresponds a location of the set of locations in the searchable code, an image of the trajectory data corresponding to the object based on the contour-coded compressed image in the database.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for identifying a spatial location of an event within video image data comprising: generating, by at least one computer device, based on trajectory data of a trajectory of an object within video image data, a searchable code that contains a set of locations traversed by the trajectory of the object within the video image, wherein the generating further comprises: dividing the video image data into a plurality of pixel regions; determining, for each pixel region in the plurality of pixel regions, whether the trajectory intersects any pixel in the pixel region; and indicating, for each pixel region, a result of the determining in a memory location in the searchable code that corresponds to the pixel region; converting, by the at least one computer device, the trajectory data into a contour-coded compressed image; and returning from a database, in response to a query having a selected location that corresponds a location of the set of locations in the searchable code, an image of the trajectory data corresponding to the object based on the contour-coded compressed image in the database.
The invention identifies where an object moved in a video. First, the video is divided into a grid of small regions. Then, for each region, the system checks if the object's path touched it. A "searchable code" is created which stores, for each region, whether the object's path went through it. The path of the object is also converted into a compressed image. Later, if someone asks "show me object paths in this region," the system looks up the region in the "searchable code." If there's a match, the corresponding compressed image of the path is retrieved from a database and displayed.
2. The method according to claim 1 , further comprising searching the database to identify a spatial location of the event within the video image data.
The method described in claim 1 (identifying where an object moved in a video by generating a searchable code based on trajectory data and a compressed image, and retrieving the image from a database in response to a query having a selected location that corresponds a location of the set of locations in the searchable code) also includes searching the database to pinpoint the location of the event (object movement) in the video.
3. The method according to claim 2 , the searching comprising: converting the area of interest to a lossy query code; and comparing the lossy query code to the searchable code of the trajectory data within the video image data.
The method described in claim 2 (identifying where an object moved in a video by generating a searchable code based on trajectory data and a compressed image, and retrieving the image from a database in response to a query having a selected location that corresponds a location of the set of locations in the searchable code and searching the database to pinpoint the location of the object movement), includes converting the area of interest into a "lossy query code" (a simplified representation of the area) and comparing this code to the "searchable code" generated from the object's path. This helps quickly find matching paths even if the query area isn't an exact match.
4. The method according to claim 3 , further comprising: decompressing the contour-coded compressed image corresponding to the lossy query code based on the comparing; and plotting the trajectory of the trajectory data within the video image data from the contour-coded compressed image.
The method described in claim 3 (identifying where an object moved in a video by generating a searchable code based on trajectory data and a compressed image, retrieving the image from a database in response to a query having a selected location, searching the database to pinpoint the location of the object movement, converting the area of interest into a "lossy query code" and comparing this code to the "searchable code" generated from the object's path) also involves decompressing the compressed image of the object's path (corresponding to the "lossy query code" match) and then displaying (plotting) the object's path onto the video.
5. The method according to claim 1 , wherein the converting further comprises: dividing the video image data into a plurality of pixel regions; determining, for each pixel region in the plurality of pixel regions, whether the trajectory intersects any pixel in the pixel region; and indicating, for each pixel region, a result of the determining in a memory location in the contour-coded compressed image that corresponds to the pixel region.
In the method described in claim 1 (identifying where an object moved in a video by generating a searchable code based on trajectory data and a compressed image, and retrieving the image from a database in response to a query having a selected location that corresponds a location of the set of locations in the searchable code), the creation of the compressed image involves dividing the video frame into small regions, checking if the object's path intersects each region, and then storing this intersection information in the compressed image itself, associating each pixel region with a corresponding memory location.
6. The method according to claim 5 , the converting further comprising: determining, for each pixel region that the trajectory intersects, a direction of the trajectory from a location in the video image associated with the pixel region; and assigning a value that indicates the direction of the trajectory from the associated location to the memory location in the contour-coded compressed image that corresponds to each pixel region that the trajectory intersects.
In the method described in claim 5 (identifying where an object moved in a video by generating a searchable code based on trajectory data and a compressed image, retrieving the image from a database in response to a query, where the compressed image generation involves dividing the video into regions and storing intersection data), after determining that the trajectory intersects a pixel region, the method also determines the direction of the trajectory within that region and stores this direction as a value in the compressed image, associating it with the corresponding region.
7. A system for identifying a spatial location of an event within video image data comprising: at least one processing device; memory operably associated with the at least one processing device; and a spatial representation tool storable in memory and executable by the at least one processing device, the spatial representation tool causing the at least one processing device to: generate, based on trajectory data of a trajectory of an object within video image data, a searchable code that contains a set of locations traversed by the trajectory of the object within the video image, wherein the generating further comprises: dividing the video image data into a plurality of pixel regions; determining, for each pixel region in the plurality of pixel regions, whether the trajectory intersects any pixel in the pixel region; and indicating, for each pixel region, a result of the determining in a memory location in the searchable code that corresponds to the pixel region; convert the trajectory data into a contour-coded compressed image; and return from a database, in response to a query having a selected location that corresponds a location of the set of locations in the searchable code, an image of the trajectory data corresponding to the object based on the contour-coded compressed image in the database.
A system for finding locations of events in video data includes a processor and memory. The memory stores software that can: create a "searchable code" that lists the regions of a video an object traveled through (by dividing the video into regions, checking each for trajectory intersection, and recording the result in memory); convert the object's path into a compressed image; and, when asked about a location, find the corresponding image of the object's path in a database using the "searchable code" and return it.
8. The system of claim 7 , the spatial representation tool further causing the at least one processing device to search the database to identify a spatial location of the event within the video image data.
The system described in claim 7 (including processor, memory, and software for creating searchable code and compressed images to locate objects in video) can also search the database to identify the location where the movement occurred.
9. The system of claim 8 , the searching comprising: converting the area of interest to a lossy query code; and comparing the lossy query code to the searchable code of the trajectory data within the video image data.
The system described in claim 8 (including processor, memory, software for creating searchable code and compressed images to locate objects in video, and database searching) performs its search by creating a simplified code representing the area of interest ("lossy query code") and comparing it with the "searchable code" of the object's path.
10. The spatial representation tool according to claim 9 further causing the at least one processing device to: decompress the contour-coded compressed image corresponding to the lossy query code based on the comparing; and plot the trajectory of the trajectory data within the video image data from the contour-coded compressed image.
The system described in claim 9 (including processor, memory, software for creating searchable code and compressed images to locate objects in video, and searching the database using a lossy query code), after finding a match using the simplified area code, can decompress the corresponding compressed image and display the object's path on the video.
11. The system according to claim 7 , wherein the converting further comprises: dividing the video image data into a plurality of pixel regions; determining, for each pixel region in the plurality of pixel regions, whether the trajectory intersects any pixel in the pixel region; and indicating, for each pixel region, a result of the determining in a memory location in the contour-coded compressed image that corresponds to the pixel region.
In the system described in claim 7 (including processor, memory, and software for creating searchable code and compressed images to locate objects in video), when creating the compressed image, the software divides the video into regions, checks if the object's path intersects each region, and then stores this intersection information in the compressed image itself, associating each pixel region with a corresponding memory location.
12. The system according to claim 11 , the converting further comprising: determining, for each pixel region that the trajectory intersects, a direction of the trajectory from a location in the video image associated with the pixel region; and assigning a value that indicates the direction of the trajectory from the associated location to the memory location in the contour-coded compressed image that corresponds to each pixel region that the trajectory intersects.
In the system described in claim 11 (including processor, memory, software for creating searchable code and compressed images to locate objects in video, where the compressed image generation involves dividing the video into regions and storing intersection data), the software, after determining that the trajectory intersects a region, determines the direction of the trajectory within that region and stores this direction as a value in the compressed image, associated with that region.
13. A computer-readable storage device storing computer instructions, which when executed, enables a computer system to identify a spatial location of an event within video image data, the computer instructions comprising: generating, based on trajectory data of a trajectory of an object within video image data, a searchable code that contains a set of locations traversed by the trajectory of the object within the video image, wherein the generating further comprises: dividing the video image data into a plurality of pixel regions; determining, for each pixel region in the plurality of pixel regions, whether the trajectory intersects any pixel in the pixel region; and indicating, for each pixel region, a result of the determining in a memory location in the searchable code that corresponds to the pixel region; converting the trajectory data into a contour-coded compressed image; and returning from a database, in response to a query having a selected location that corresponds a location of the set of locations in the searchable code, an image of the trajectory data corresponding to the object based on the contour-coded compressed image in the database.
A computer-readable storage medium (like a hard drive or flash drive) contains instructions that, when run, allow a computer to find locations of events in video data. The instructions cause the computer to: create a "searchable code" that lists the regions of a video an object traveled through (by dividing the video into regions, checking each for trajectory intersection, and recording the result in memory); convert the object's path into a compressed image; and, when asked about a location, find the corresponding image of the object's path in a database using the "searchable code" and return it.
14. The computer-readable storage device according to claim 13 , further comprising computer instructions for searching the database to identify a spatial location of the event within the video image data, the searching comprising: converting the area of interest to a lossy query code; and comparing the lossy query code to the searchable code of the trajectory data within the video image data.
The computer-readable storage medium described in claim 13 (containing instructions for creating searchable code and compressed images to locate objects in video) also includes instructions for searching the database to identify the location of the event, by creating a simplified code representing the area of interest ("lossy query code") and comparing it with the "searchable code" of the object's path.
15. The computer-readable storage device according to claim 14 , further comprising computer instructions for: decompressing the contour-coded compressed image corresponding to the lossy query code based on the comparing; and plotting the trajectory of the trajectory data within the video image data from the contour-coded compressed image.
The computer-readable storage medium described in claim 14 (containing instructions for creating searchable code and compressed images to locate objects in video, and searching the database using a lossy query code), also includes instructions for decompressing the compressed image corresponding to a matching "lossy query code" and displaying the object's path on the video.
16. The computer readable storage device according to claim 13 , wherein the converting further comprises: dividing the video image data into a plurality of pixel regions; determining, for each pixel region in the plurality of pixel regions, whether the trajectory intersects any pixel in the pixel region; and indicating, for each pixel region, a result of the determining in a memory location in the contour-coded compressed image that corresponds to the pixel region.
In the computer readable storage device described in claim 13 (containing instructions for creating searchable code and compressed images to locate objects in video), the instructions for creating the compressed image involve dividing the video into small regions, checking if the object's path intersects each region, and then storing this intersection information in the compressed image itself, associating each pixel region with a corresponding memory location.
17. The computer readable storage device according to claim 16 , the converting further comprising: determining, for each pixel region that the trajectory intersects, a direction of the trajectory from a location in the video image associated with the pixel region; and assigning a value that indicates the direction of the trajectory from the associated location to the memory location in the contour-coded compressed image that corresponds to each pixel region that the trajectory intersects.
In the computer readable storage device described in claim 16 (containing instructions for creating searchable code and compressed images to locate objects in video, where the compressed image generation involves dividing the video into regions and storing intersection data), the instructions also include determining, for each region the trajectory intersects, the direction of the trajectory and storing this direction as a value in the compressed image, associated with that region.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
August 22, 2016
August 8, 2017
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.